Pulse code modulation system suitable for digital recording of broadband analog signals

Abstract

The d.c. component of video signals is to a great extent eliminated for a wide range of content, sufficiently for recording, for example, without the introduction of redundancy in the digital coding, when the digital words corresponding to samples of the analog signal are converted into a binary code in which the number of bits of one logic level is approximately linearly dependent upon the magnitude of the analog signal. In such a code, there are successive ranges of the analog signal in which the number of bits of one logic level remains the same, and within said ranges the change in digital words from one analog level to the next involves the change of two bits of the word. The digital words are serially provided for recording or transmission.

I claim: 
     
       1. Pulse code modulation method for use in recording digital signals or transmitting them through a channel having a high-pass characteristic, comprising the steps of: converting an analog signal to be recorded or transmitted into a sequence of digital words, which respectively represent quantized samples of the analog signal level in a binary code at a sampling rate not lower than twice the highest frequency in said analog signal;   inverting alternate digital words, each corresponding to a quantized sample provided as the result of analog-to-digital conversion, and   converting said digital words into coded words, such that within each said coded word the number of bits having a selected one of the two states, 1 or 0, has an approximately linear dependence upon the value of the corresponding analog signal level.   
     
     
       2. A method as defined in claim 1, in which within a level range of the analog signal for which said coded words have the same number of bits of said one state, the coded words change from one quantized level to an adjacent quantized level of the analog signal, in each case, by a change of two bits of the coded word. 
     
     
       3. A method as defined in claim 1 or claim 2, in which the analog signal has its frequency band limited before being supplied to the analog-to-digital converter. 
     
     
       4. A method as defined in claim 3, in which the coded words are 8-bit words and in which the change from one quantized analog signal level to an adjacent quantized analog signal level involves the change of not less than one bit and not more than two bits of the coded word. 
     
     
       5. Apparatus for encoding analog signals digitally for recording or for a transmission channel having a high-pass characteristic, comprising: means for converting said analog signal to a digital words, respectively representing quantized levels of samples of said analog signals at a sampling rate which is not less than twice the highest frequency of said analog signal;   means for inverting alternate digital words of the output of said converting means, and   code conversion means for putting the digital words representing each analog sample into coded words into which the number of bits having a selected one of the two states, 1 or 0, has an approximately linear dependence upon the corresponding analog signal level.   
     
     
       6. Apparatus as defined in claim 5, in which said code conversion means is constituted to convert said digital words into coded words in which there are level ranges of the analog signal for which the number of bits of said one of said two states, 1 or 0, remains constant and in which code, within said level ranges with a constant number of bits of said one of said two states, the change of coded words from one quantized level of said analog signal to an adjacent quantized level of said analog signal is a change of two bits of the coded word. 
     
     
       7. Apparatus as defined in claim 6, in which said coded words are 8-bit words and the change of said 8-bit words, in going from one quantized analog signal level to an adjacent quantized analog signal level, is a change of either one or two bits in every case. 
     
     
       8. Apparatus as defined in claim 5,6 or 7, in which a low-pass filter for limiting the frequency range of said analog signal is provided ahead of said analog-to-digital converter. 
     
     
       9. Apparatus as defined in claim 5,6 or 7, in which said analog signal is a video signal comprising synchronizing signals, and in which a synchronizing signal generator is provided, as well as means for synchronizing said generator to the synchronizing signals of said video signal, and in which said synchronizing signal generator has a clock frequency output to control the sampling rate of the digital-to-analog converter. 
     
     
       10. Apparatus as defined in claim 5,6 or 7, in which parallel-to-serial digital signal conversion means are provided for converting the output of said code conversion means to a serial digital signal. 
     
     
       11. A method of decoding a digital signal representing an analog signal and encoded in a code in which each coded word representing a quantized sample of the analog signal has a number of bits of a selected one of the two states of the binary code which number depends approximately linearly upon the level of the corresponding quantized analog signal sample, comprising the steps of: converting the coded words of said digital signals so as to put each coded word thereof into digital words readily converted into analog signals;   inverting alternate digital words resulting from the foregoing conversion step, and   converting the resulting digital words into quantized analog signal samples and combining said samples to provide an analog signal.   
     
     
       12. A method as defined in claim 11, in which the coded words are in serial form and are subjected to a serial-to-parallel conversion before the coded word conversion step. 
     
     
       13. Apparatus for decoding digital signals composed of digital words, each representing a quantized sample of an analog signal and each encoded in coded words in which the number of bits of a selected one of the two states, 1 or 0, of the binary code in each word has an approximately linear dependence upon the level of the corresponding quantized analog signal sample, comprising: code conversion means for putting the coded words into a digital words more readily convertible into an analog signal;   means for inverting alternate digital words of the output of said code conversion means;   means for converting the digital words provided by the output of said code conversion and inverting means into analog signal samples, and   means for combining said analog signal samples to provide an analog signal.   
     
     
       14. Apparatus as defined in claim 13, in which said code conversion means is constituted to operate on an input coding in which there are magnitude ranges of the analog signal samples represented by said coded words for which the number of bits per word of said one of said two states remains constant and in which code, within said magnitude ranges for a constant number of bits of said one of said two states in a word, the difference between words representing adjacent quantized magnitude levels of analog signal samle magnitude is a change of two bits of the word. 
     
     
       15. Apparatus as defined in claim 14, in which the coded words of said digital signals are 8-bit words and the difference between 8-bit words, before code conversion, representing adjacent analog signal sample quantized magnitude levels is a chage of either one or two bits in every case. 
     
     
       16. Apparatus as defined in claim 14 or 15, in which the coded words are in a serial stream initially, and in which means for serial-to-parallel conversion of said coded words for providing parallel bit transmission of data words thereof to said code conversion means are connected ahead of said code conversion means. 
     
     
       17. Pulse code modulation method for use in recording digital signals or transmitting them through a channel having a high-pass characteristic, comprising the steps of: converting the analog signal to be recorded or transmitted into digital words at a sampling rate not lower than twice the highest frequency in said analog signal;   converting the digital words into coded words, such that within each said coded word the number of bits having a selected one of the two states of the binary code has an approximately linear dependence upon the level of the corresponding analog signal, and   inverting alternate coded words produced by the step of converting digital words into coded words.   
     
     
       18. Pulse code modulation method for use in recording digital signals or transmitting them through a channel having a high-pass characteristic, comprising the steps of: converting the analog signal to be recorded or transmitted into digital words at a sampling rate not lower than twice the highest frequency in said analog signal,   converting the digital words into coded words, such that amplitude values being symmetrical to the 50% level of the entire amlitude range are represented by coded words which are inverse to each other, and that   within each said coded word the number of bits having a selected one of the two states, 1 or 0, of the binary code has an approximately linear dependence upon the level of the corresponding analog signal, and then inverting alternate coded words produced by the step of converting digital words into coded words.